Separating Atrial Flutter From Atrial Fibrillation With Apparent Electrocardiographic Organization Using Dominant and Narrow F-Wave Spectra
Bobbi L. Hoppe, MD,
Andrew M. Kahn, MD, PhD,
Gregory K. Feld, MD, FACC,
Alborz Hassankhani, MD, PhD and
Sanjiv M. Narayan, MB, MD, FACC*
Electrophysiology Service, Veterans Affairs San Diego, University of California San Diego, San Diego, California.
View larger version (36K):
[in a new window]
|
Figure 1 Extracting atrial waveform. (A) In this electrocardiogram (ECG) of typical atrial flutter (AFL), the labeled F-wave template is correlated to successive ECG time points, yielding (B) a correlation time series ("filtered atrial waveform"). Peaks (r approximately equal to 1) identify recurrent F waves even if they overlap QRS/T waves. Lead II is shown in panel A to depict sawtooth F waves, but only leads V5 or I (X-axis), aVF (Y-axis), and V1 (Z-axis) were analyzed in this study.
|
|
View larger version (62K):
[in a new window]
|
Figure 2 (A) An ECG of atypical AFL coded as atrial fibrillation (AF). Intracardiac tracings confirmed right atrial (RA) AFL. (B) Correlation-time series show regularity in "filtered atrial ECGs" (particularly V1). (C) Correlation spectra show dominant peaks at 4.04 Hz that are tall and narrow (area ratios = 0.51, 0.56, and 0.81 in X-, Y-, and Z-axes, respectively; 2.5-Hz bandwidth). Other abbreviations as in Figure 1.
|
|
View larger version (53K):
[in a new window]
|
Figure 3 Electrocardiographic and intracardiac tracings of (A) atypical AFL and (B) "organized" AF, each showing atypical and varying F waves. (Middle panels) correlation series of each template to its ECG. (Bottom panels) spectra from 213 points (4 s shown) show a dominant peak (A) at 3.2 Hz that is narrow, with area ratio of 0.48 (2.5-Hz bandwidth), and tall (relative height 9.2 dB), suggesting a single AFL circuit; (B) at 6.07 Hz (3- to 10-Hz window) that is broader (area ratio 0.33) and shorter (3.7 dB), suggesting multiple AF wave fronts with mean pulmonary vein (PV) cycle length (CL) predicted by ECG-derived frequency of 6.07 Hz. Other abbreviations as in Figures 1 and 2.
|
|
View larger version (29K):
[in a new window]
|
Figure 4 (A) Relative peak heights were higher in AFL than AF (p < 0.001), yet resulted in three misclassifications at the optimal cutpoint ( 7.89 dB). (B) Spatial non-uniformities in relative peak heights for each patient were not consistent for either group (Table 2). (C) Receiver operating characteristic (ROC) (area 0.94). Other abbreviations as in Figures 1 and 2.
|
|
View larger version (29K):
[in a new window]
|
Figure 5 (A) Peak area ratio (three-axis mean) to varying envelope widths show that peak area ratios fall as envelopes widen from 0.625 to 5 Hz, although differences between AF and AFL were maintained (Table 2). Small icons represent individual patients and large icons represent the mean (± SD) for each group. (B) Receiver operating characteristic (ROC) curves for the diagnosis of AFL at each envelope width. The arrow indicates the optimum envelope width (2.5 Hz; Fig. 6). Other abbreviations as in Figures 1 and 2.
|
|
View larger version (29K):
[in a new window]
|
Figure 6 (A) Optimum peak area ratio (envelope width 2.5 Hz) separated all AFL from AF patients using cutpoint 0.44 (three-axis mean). (B) Receiver operating characteristic (ROC) area under the curve was, thus, better for 2.5 Hz (area = 1) than other envelopes. (C) Between spatial axes, 2.5-Hz peak area ratios varied for each patient but not consistently for either group (Table 2). Other abbreviations as in Figures 1 and 2.
|
|
View larger version (37K):
[in a new window]
|
Figure 7 Diagnostic examples. (A) Analysis suggests AFL, with peak area ratio 0.53 and relative height 14.5 dB at 4.03 Hz. Atypical AFL was confirmed at EPS, and ablated near superior vena cava (SVC) baffle. (B) Electrocardiographic analysis suggested AFL with peak area ratio 0.82 and relative height 19.8 dB at 5.49 Hz. Atypical AFL was ablated in the anterior left atrium. (C) Analysis confirmed AF, with relative peak height 3.46 dB and area ratio = 0.41 at 8.67 Hz. (D) Spectral peak (6.59 Hz) was misclassified as AFL by relative height (10.6 dB), but area ratio = 0.29 assigned the correct diagnosis of AF (Patient #30; see fractionated and irregular electrograms). For all panels, arrows reflect 2.5-Hz envelope, and 8.192 s were analyzed (4 s are shown). Other abbreviations as in Figures 1 and 2.
|
|
|
