CORRESPONDENCE: LETTER TO THE EDITOR
Reply
David G. Benditt, MD, FACC, FHRS, FRCPC*,
Demetris Yannopoulos, MD,
Keith G. Lurie, MD, FACC,
Scott Sakaguchi, MD, FACC, FHRS,
Laura Van Heel, RN,
Simon Milstein, MD, FACC and
Cengiz Ermis, MD
* University of Minnesota Medical School, Cardiac Arrhythmia Center, 420 Delaware Street SE, MMC 508, Minneapolis, Minnesota 55455 (Email: bendi001{at}umn.edu).
Drs. Adelstein and Saba raise a number of criticisms regarding our recent publication in which we observed that the number of atrial tachyarrhythmia (AT) events in heart failure patients diminished after conventional dual-chamber pacemakers or defibrillators were "upgraded" to cardiac resynchronization (CRT) systems (1). We have endeavored to address their comments.
First, Drs. Adelstein and Saba indicate disappointment in our not citing their publication (2). This omission was not intentional; our manuscript was accepted on April 24, 2007, and was at the publishers when theirs was printed (July 15, 2007). Thus, publication schedules crossed. Second, they criticize how we counted AT events (i.e., mode switch counts with electrogram review when available) without having documentation of all events (a limitation noted by us). They are concerned that transient undersensing may result in our having counted "multiple" apparent individual AT episodes (instead of 1 ongoing episode), thereby overestimating AT frequency; further, they suggest that, absent electrograms, we could not substantiate that ATs were present. With regard to the former, we believe that the device algorithms used and our measurement techniques minimized the "overcounting" problem. However, if, as they contend, it was impossible to eliminate all "overcounting," our "upgrade study model," with each patient being his/her own control, inherently protected against an important counting bias; error susceptibility would be similar before and after CRT upgrade. In this regard, our study design is more robust than their case-control methodology (2). With regard to the second aspect of their documentation criticism, one might be led to assume that in their report Drs. Adelstein and Saba either vetted every mode switch event with electrograms or censored those that could not be vetted. However, neither seems to be true. In fact, although they indicate that "electrographic data" were available in 88% of patients, they do not state unequivocally that all events were documented by electrograms. The probability of complete vetting is unlikely unless either data downloads were very frequent or undocumented events were censored (which could result in undercounting). In fact, they admit that absent electrocardiographic data "... episode counters and mode switch durations ... were assumed to be correct." In effect, their study is subject to the same criticism they level at us, while lacking the internal controls offered by our study design. Third, Drs. Adelstein and Saba criticized our focusing on AT events and AT rate as clinical end points, believing that AT burden is more important. In this regard, we believe that episode duration is an important determinant of thromboembolic risk, but otherwise our opinions are similar (3). Indeed, whereas the number of AT events and durations that we provided may not be a precise "burden" measure, they provide a "burden" surrogate. Finally, the Pittsburgh researchers point out that, unlike both Fung et al. (4) and our study (1), they did not observe a significant CRT benefit. However, that may have been due to limitations of their study design. In fact, their figures suggest that CRT patients did better than control patients; specifically, CRT was associated with a trend toward greater AT-free proportion at any given time. Potentially, had their population been larger, or had they used a more robust study model (such as our "upgrade" approach), the trend toward benefit may have been significant.
Differences aside, we appreciate the interest shown by Drs. Adelstein and Saba in our study and look forward to their continuing to contribute to the better understanding of CRT in clinical care.
 |
References
|
|---|
- Yannopoulos D, Lurie KG, Sakaguchi S, et al. Reduced atrial tachyarrhythmia susceptibility after upgrade of conventional implanted pulse generator to cardiac resynchronization therapy in patients with heart failure J Am Coll Cardiol 2007;50:1246-1251.[Abstract/Free Full Text]
- Adelstein EC, Saba S. Burden of atrial fibrillation after cardiac resynchronization therapy Am J Cardiol 2007;100:268-272.[CrossRef][ISI][Medline]
- Ermis C, Seutter R, Zhu AX, et al. Impact of upgrade to cardiac resynchronization therapy on ventricular arrhythmia frequency in patients with implantable cardioverter defibrillators J Am Coll Cardiol 2005;46:2258-2263.[Abstract/Free Full Text]
- Fung JW-H, Yo C-M, Chan JY-S, et al. Effects of cardiac resynchronization on incidence of atrial fibrillation in patients with poor left ventricular systolic function Am J Cardiol 2005;96:728-731.[CrossRef][ISI][Medline]
Related Article
-
Cardiac Resynchronization Therapy and Atrial Tachyarrhythmias: A Question Still Searching for an Answer
- Evan Adelstein and Samir Saba
J. Am. Coll. Cardiol. 2008 51: 676-677.
[Full Text]
[PDF]
Top
References