reply: We thank Dr. J. E. Madias (3) for his perceptive comments on our article, “High-resolution electrical mapping of depolarization and repolarization alternans in an ischemic dog model” (1). Our response to his queries are as follows. The first query was, “Since electrical alternans (EA) was also seen before ischemia, is it not a contradiction that the presence, and not the amplitude, of EA is predictive of ventricular fibrillation (VF)?” The results of our study suggest that there may be a critical amount of area with EA that is necessary for spontaneous VF to develop. Although there were low-level alternans during the preischemic period, none of the dogs had alternans prevalence above the cutoffs determined in our receiver-operator characteristic curve analysis for the respective measurements. This is consistent with the fact that none of the dogs spontaneously developed VF during the preischemia period.
His second query was, “Could the methodology used be implemented in standard ECGs or Holter monitorings?” The answer is yes. This methodology can be used to quantify alternans amplitude and provide an amplitude-to-error ratio in standard ECGs and Holter recordings. It would have the potential advantage of a higher time resolution than the spectral or modified moving average techniques. However, the trade-off is the lack of noise reduction with this technique. It would have to be determined what amplitude-to-error ratios are necessary for our method to have predictive value when analyzing microvolt alternans in standard ECGs or Holter recordings.
Dr. Madias points out an interesting observation in his third query: the presence of S-wave alternans in Fig 3. Alternans in the S wave was commonly seen in the signals that had clear S waves. As alluded to in our limitation section, the S wave of the unipolar electrogram likely includes both a late depolarization and an early repolarization component. Therefore, it is unclear whether the S-wave alternans is a result of depolarization or repolarization alternans. This feature of the S wave may be an interesting topic for further study.
Regarding Dr. Madias's questions and comments about correlation between waveform amplitude and alternans amplitude and the use of scaling, all the electrograms were scaled by the peak-to-peak QRS to account for amplitude differences in the electrograms that may be due to nonphysiological reasons such as differences in electrode contact quality. All measurements were made off the scaled electrograms. The ST/T wave was not scaled separately from the QRS, but this method as shown in Dr. Madias's article (2) may be useful for future analysis of the T wave since it is likely that there is at least some correlation between waveform amplitude and alternans amplitude. We thank Dr. Madias for this suggestion.
No conflicts of interest are declared by the author(s).
- Copyright © 2010 the American Physiological Society