Author + information
I would like to offer 2 comments on the systematic review by Ganesan et al. (1) and the accompanying editorial by Gasparini and Galimberti (2). First, the review suggests that atrioventricular nodal ablation (AVNA) in patients who have atrial fibrillation (AF) and biventricular pacing may reduce all-cause mortality. Calculations based on the studies summarized in Tables 1 and 2 (1) showed that all-cause mortality in patients who had undergone AVNA was 19 of 189 (about 10%) and 65 of 261 (about 25%); that is, the number needed to treat (NNT) was approximately 7 over a variable median follow-up of 6 to 34 months. These results are indeed striking, particularly compared with results of randomized trials of biventricular pacing in sinus rhythm versus medical therapy alone (245 of 1,472 [about 17%] vs. 231 of 1,129 [about 20%]; that is, a 3% absolute reduction in all-cause mortality (3). Even at 3 years, the NNT to prevent 1 death was 9 based on the data from the Cardiac Resynchronization–Heart Failure study (4).
Although it is possible that the benefit of increasing the percent biventricular pacing in patients with AF from <85% (the criterion for AVNA in the largest study included in the review) to close to 100% with AVNA may exceed that of patients in sinus rhythm, it is more plausible that the benefit of AVNA reported in this review is an exaggeration of the “true” benefit. Nonrandomized studies, particularly retrospective studies, are more likely to show favorable results with larger treatment effects (5). In this regard, it is notable that all 3 studies which reported on mortality as an outcome in this review were retrospective in design, and 2 of the 3 were single-center studies.
Second, the true safety of AVNA in patients who have heart failure and AF should be assessed over the longer term (not simply procedural safety) because it results in irreversible pacemaker dependency. Short-term follow-up may not capture the long-term risks of an irreversible procedure. What are the long-term risks, therefore, of pacemaker dependency? The maximum follow-up in the studies included in the review by Ganesan et al. (1) was a median of 34 months, which is less than the lifetime of a pacemaker/defibrillator generator. With significant risks associated with generator replacement/revisions (particularly in patients with biventricular pacemakers/defibrillators) (6), the clinical implications of irreversible pacemaker dependency cannot be easily dismissed. Furthermore, in patients who are less likely to respond to biventricular pacing (e.g., relatively narrow QRS duration and extensive areas of infarction), does the potential benefit of AVNA still outweigh the potential complications of irreversible pacemaker dependency?
Hence, the current picture of AVNA in heart failure is incomplete. The need for randomized studies is clear.
- American College of Cardiology Foundation
- Ganesan A.N.,
- Brooks A.G.,
- Roberts-Thomson K.C.
- Gasparini M.,
- Galimberti P.
- Poole J.E.,
- Gleva M.J.,
- Mela T.,
- et al.