Author + information
- Sherif F. Nagueh, MD⁎ ()
Reprint requests and correspondence:
Dr. Sherif F. Nagueh, Methodist DeBakey Heart and Vascular Center, 6550 Fannin Street, SM-677, Houston, Texas 77030
Cardiac resynchronization therapy (CRT) is an important treatment modality for patients with advanced heart failure (1). Although the clinical benefits of CRT are undisputed, its high cost, potential complications, and the 20% to 30% rate of nonresponders make it essential to achieve better case selection, rather than depending on the QRS duration only.
Cardiac imaging plays an important role in the evaluation of patients with heart failure, including providing data on mechanical dyssynchrony (2). In particular, a number of observational studies have shown that tissue Doppler and speckle tracking can be used to diagnose mechanical dyssynchrony and identify likely responders to CRT (1). However, there remains a need for a methodology that is more accurate and reproducible. Accordingly, few investigators have evaluated 3-dimensional (3D) echocardiography for imaging dyssynchrony. The published reports have shown that 3D echocardiography is a technique that holds good potential (3–5). The study by Sonne et al. (6) in this issue of iJACC is another look at 3D echocardiography that provides fundamental data that are critical to the ongoing evaluation of this technique for imaging dyssynchrony. To help appreciate its implications, one needs to answer 2 basic questions: what constitutes a perfect technique for evaluating dyssynchrony? How close does 3D echocardiography come to the perfect technique?
The Perfect Technique
The ideal imaging modality should have high temporal and spatial resolution, high signal-to-noise ratio, and low interobserver variability. The technique should be readily available, user friendly, and cost effective. In addition, for imaging the CRT population, it should allow satisfactory and safe imaging in the presence of pacemakers and defibrillators. In considering the above attributes, one can accept sophisticated techniques if they prove to be safe, highly accurate, and reproducible. Furthermore, based on the history of research in this field, the results of any promising technique should be verified in multicenter studies (the individual sites should still show satisfactory quality control before participating in multicenter research endeavors).
Before delving into a critique of using 3D echocardiography to image dyssynchrony, one needs to review a few key technical points about 3D in general. The earliest 3D techniques were based on the acquisition of multiple 2-dimensional (2D) images, followed by reconstruction into a 3D dataset. These needed gating and were time consuming in their acquisition and analysis. The more recent systems allow for true real-time acquisition with better spatial resolution and with faster online analysis. However, they are still dependent on image quality, at least to the same extent as 2D, and importantly, from the perspective of dyssynchrony imaging, they have the limitations of low frame rate and gated capture.
3D Echocardiography for Dyssynchrony Imaging: The Current Study
In the study by Sonne et al. (6), 3D echocardiography was used to acquire 4 wedge-shaped smaller volumes from which a single volume data set was derived. Segmental volume–time curves were generated for 16 or 17 segments. The time to the smallest segmental volume was measured for each of the segments, and the standard deviation (SD) of that measurement in all segments was used as the dyssynchrony index. The SD approach is similar to that of other published reports (3) that used 3D echocardiography for analysis of dyssynchrony.
Based on the above, one can readily predict that satisfactory endocardial border definition is essential throughout the cardiac cycle. In addition, because the index is essentially an SD, it is heavily affected by extremes. This in and of itself renders it susceptible to higher variability and major changes based on the quality of data from any 1 segment.
Unlike previous reports, the article by Sonne et al. (6) included a large number of normal individuals in whom the 3D dyssynchrony index was determined. This is always an important first step in trying to reliably obtain narrow confidence intervals for diagnosing dyssynchrony. The investigators confirmed the expected result of a larger SD with the inclusion of more segments, albeit a small difference in normal hearts (4% in a 16-segment model vs. 4.5% in a 17-segment model). Based on the cutoff derived from the normal group, all patients with dilated cardiomyopathy had dyssynchrony. The investigators raised the question of whether this is a real observation or a result reflective of technical limitations and subsequent errors. The latter conclusion is favored because one would not expect all patients with dilated cardiomyopathy and a normal QRS duration to have dyssynchrony.
How Does 3D Compare With the Perfect Technique?
In the evaluation of dyssynchrony, 3D echocardiography is readily available and certainly is safe to use in patients with implanted pacemakers and defibrillators. However, it has a number of important limitations that were exposed by the important study by Sonne et al. (6). Specifically, its low frame rate (usually 20 to 30 frames/s) results in low temporal resolution. Furthermore, the segmental volume curves are constructed not from a single cardiac cycle but from 8. There is also the challenge of determining the minimum volume from curves with exceedingly low amplitudes in patients with severely depressed ejection fraction, which remains an issue even with adequate endocardial border definition.
A major problem is also apparent from the results of the study. Intraobserver variability could be as high as 21%, whereas interobserver variability is even larger. Because these results were obtained from an expert laboratory, the technique is clearly not ready for clinical application based on this observation alone.
Aside from the above major limitations, the concept of recognizing delay between different segments is not specific for a response to CRT in patients with coronary artery disease. In the latter group of patients, ischemia and/or infarction can be the reason for the observed delays, and their treatment may be the more prudent approach rather than CRT. This is not a problem unique to 3D echocardiography, but for all imaging techniques in general. In summary, one can conclude that the application of 3D echocardiography with current technology to image dyssynchrony is not advocated at present, but the topic is not closed yet.
What Can 3D Offer Patients With or Undergoing Evaluation for CRT?
There are a number of clinically useful applications for 3D echocardiography in this population. These include the accurate measurement of left ventricular volumes and ejection fraction (7). The latter is a critical piece in the selection criteria for CRT. The consensus from the published literature supports higher accuracy for 3D echocardiography, versus 2D, when compared with cardiac magnetic resonance as the gold standard. Further, based on its higher accuracy, 3D can better track changes in left ventricular volumes and ejection fraction after CRT, which is of particular interest because left ventricular volumes predict outcome much better than clinical status after biventricular pacing (8). The ability to present visually appealing images of lead location and the relation of the lead to areas of regional dysfunction could also be helpful for the CRT team.
It is likely that future developments with improvements in temporal and spatial resolution will bring 3D echocardiography closer to successful imaging of dyssynchrony. In particular, the ability to acquire a complete 3D dataset in 1 cardiac cycle can have a major impact by reducing artifacts caused by respiration and cardiac translation. As pointed out by Sonne et al. (6), it is essential to explore other analytic approaches (other than SD) to achieve this ambitious and clinically rewarding goal for 3D echocardiography.
Dr. Nagueh served as consultant to St. Jude Medical and GE in 2008.
↵⁎ Editorials published in JACC: Cardiovascular Imaging reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Imaging or the American College of Cardiology.
- American College of Cardiology Foundation
- Gorcsan J. III.,
- Abraham T.,
- Agler D.A.,
- et al.
- Nagueh S.F.
- Kapetanakis S.,
- Kearney M.T.,
- Siva A.,
- et al.
- Liu W.H.,
- Chen M.C.,
- Chen Y.L.,
- et al.
- Sonne C.,
- Sugeng L.,
- Takeuchi M.,
- et al.
- Yu C.M.,
- Bleeker G.B.,
- Fung J.W.,
- et al.