Author + information
- ↵⁎Reprint requests and correspondence:
Dr. Anne B. Curtis, University at Buffalo, Department of Medicine, 219 D.K. Miller Building, Erie County Medical Center, 462 Grider Street, Buffalo, New York 14215
Clinical trials of cardiac resynchronization therapy (CRT) have consistently demonstrated significant improvement in quality of life, functional status, and exercise capacity in patients with heart failure (HF). CRT is also associated with reverse left ventricular (LV) remodeling, improved systolic and diastolic function, and decreased mitral regurgitation. In addition, improvement in survival was reported in the COMPANION (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure) and the CARE-HF (Cardiac Resynchronization–Heart Failure) trials (1).
Despite the positive outcomes observed with CRT in HF, up to 30% of patients show no improvement with CRT. Such failure to improve may be due to a lack of baseline LV dyssynchrony, the presence of substantial scar tissue, suboptimal LV lead position, and/or uncorrected dyssynchrony. This residual dyssynchrony can occur at multiple levels: in the atria, between the atria and the ventricles, or between the ventricles. Multiple methodologies have been proposed to reduce cardiac dyssynchrony and optimize CRT response through programming of the atrioventricular (AV) and ventricle-ventricle intervals, but no consensus has been reached on which method is superior. Interestingly, the approach to AV optimization varied widely in the CRT clinical trials, with no AV optimization at all in the CONTAK CD trial, suggesting that even when imperfectly applied, a substantial proportion of patients can derive clinical benefit from CRT.
In this issue of iJACC, Liang et al. (2) elegantly describe the acute hemodynamic effects of atrial sensing (AS) versus atrial pacing (AP) in CRT patients and find that AS is the optimal mode of programming in CRT. We would like to emphasize 3 main findings in this study. First, almost all Doppler-based measures of ventricular hemodynamic performance, diastolic filling time, and global strain were significantly better in AS compared with AP mode. Second, in AS mode, active atrial strain (atrial contractility) was significantly higher, and all measurements of intra-atrial time delays (mechanical synchrony) were significantly shorter. These findings indicate that the acute hemodynamic benefits of AS are probably due to improved atrial contractility and synchrony and suggest that AP could lead to atrial dyssynchrony. Third, the optimal AV delay (AVD) observed with AP mode was ≥30 ms longer than the optimal AVD with AS mode in 65% of the patients, indicating that a routine fixed paced offset of 0 to 30 ms might be inadequate for the majority of CRT patients. Interestingly, Gold et al. (3) compared 3 noninvasive methods (Doppler based, intracardiac electrogram based, and fixed AVD) with invasive measurements of LV dP/dt to determine the optimal AVD during AS and AP modes in 28 patients undergoing CRT. The investigators found that the optimal AVD in AP mode was associated with a significantly greater increase in LV dP/dt compared with AS mode. They also found that the optimal AVD with the intracardiac electrogram based method had better correlation with the maximum achievable LV dP/dt compared with Doppler-based or fixed AVD techniques. Similar to the study by Liang et al. (2), the optimal AVD observed with AP mode was ≥30 ms longer than the optimal AVD with AS mode in 88% of their patients, confirming that a fixed paced offset of 0 to 30 ms might be inadequate for the majority of CRT patients. The discrepancies in the findings between the study by Gold et al. (3) and the present study are difficult to explain, because both studies had adequate methodologies. Hence, further clinical studies are required to clearly establish the optimal method to assess hemodynamic improvement in CRT patients.
Data on the deleterious effects of long-term AP have been overlooked in the clinical trials evaluating pacing therapy. These trials had mainly focused on the effects of ventricular dyssynchrony induced by right ventricular pacing. From these trials, we learned that >40% cumulative percent ventricular pacing is associated with a higher incidence of HF and atrial fibrillation. These trials also showed that the adverse outcomes of ventricular dyssynchrony induced by right ventricular pacing appear to be time dependent and modulated by baseline LV systolic function, with an earlier onset in patients with depressed LV systolic function. Moreover, these adverse clinical outcomes have been associated with concomitant atrial and ventricular remodeling (1). Hence, it is also possible that a high long-term cumulative percent AP could be associated with further worsening in atrial function.
Programming AS mode in CRT patients may be difficult to achieve in those with concomitant sinus node dysfunction (SND) and/or chronotropic incompetence. Although the prevalence of SND in HF is not well known, SND is often observed in patients with HF. HF is associated with significant remodeling of the sinus node, prolongation of corrected sinus node recovery times, and abnormal propagation of the sinus impulse. Moreover, SND could further be worsened by medications with negative chronotropic effects, such as beta-blockers, amiodarone, and digoxin, which are commonly prescribed in patients with HF. At present, it is not clear if lowering the lower rate limit for pacing to 50 beats/min would result in higher AS rates and hence better hemodynamic status.
It is important to recognize that the optimal AVDs at rest and during exercise are different. Mokrani et al. (4) found that the optimal AVD on the basis of the LV outflow tract velocity-time integral was shorter during exercise than at rest in 37%, unchanged in 37%, and longer in 26% of patients. Optimization of the AVD during exercise increased the LV outflow tract velocity-time integral significantly compared with any fixed AVD, optimal AVD at rest, and systematic shortening of AVD during exercise (rate-adaptive AVD algorithm).
Clinical studies have demonstrated that fusion or pseudo-fusion decreases the long-term efficacy of CRT therapy. For this reason, the AS and AP AVDs are typically programmed short to avoid fusion and pseudo-fusion, because correction of LV dyssynchrony requires that biventricular paced activation replace native ventricular activation. Hence, lengthening of the AP AVD to allow AS AVD in CRT patients could create competition between continuous biventricular pacing and native ventricular activation.
The results of the SMART-AV (Comparison of AV Optimization Methods Used in Cardiac Resynchronization Therapy) trial add further complexity to our understanding of CRT optimization. In this trial, 980 CRT patients were prospectively randomized to a fixed AVD, an echocardiographically optimized AVD, or an AVD optimized with SmartDelay (an electrogram-based algorithm [Boston Scientific Corporation, Natick, MA]). The primary end point was LV end-systolic volume. Secondary end points included New York Heart Association functional class, quality-of-life score, 6-min walking distance, LV end-diastolic volume, and LV ejection fraction. No differences in primary or secondary end points were observed at 6 months.
The ideal sensed or paced AVD in CRT will probably need to be adjusted automatically and dynamically to adapt to constant changes in hemodynamic conditions, which could be different for each patient. A recent study suggested that the insertion of a sensor in the pacing lead could allow more dynamic and automatic adjustments of the optimal AS or AP AVD (5).
In summary, programming to optimize the delivery of CRT remains challenging. Future clinical trials should focus on the long-term clinical benefits of hemodynamic sensors embedded in pacing leads that will automatically and periodically change the AVD (sensed or paced) to better adjust to the constant changes in patients' hemodynamic needs, which can vary widely from time to time and from patient to patient. Clinical trials that compare the results with hemodynamic sensors versus current echocardiography-based optimization techniques are desirable. We also need clinical investigations into the benefits of long-term AP versus AS. The latter are complicated by practical issues of underlying sinus bradycardia. Separating out the contributions of rate versus method of atrial activation, sensed versus paced, will be challenging. In any event, clinical outcomes are the preferred outcome measure, rather than simple echocardiographic indexes of response. In programming CRT, we should remember: don't forget the atrium!
Dr. Curtis is a consultant and a member of the Speakers' Bureau for and has received research sponsorship from Medtronic, Inc., is a consultant for and has received research sponsorship from St. Jude Medical, is a consultant for Biosense Webster, and is a consultant and member of the Speakers' Bureau for Sanofi-Aventis. Dr. Lopera has reported that he has no relationships to disclose.
↵⁎ 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
- Lopera G.A.,
- Curtis A.B.
- Liang H.-Y.,
- Cheng A.,
- Chang K.-C.,
- et al.