April 06, 2011

April 6, 2011 (New Orleans, Louisiana) — Using speckle-tracking echocardiography to guide placement of left ventricular pacing leads can reverse remodeling and avert some deaths or heart-failure hospitalizations in patients implanted with cardiac resynchronization therapy (CRT) devices, a randomized trial has suggested.

Dr Fakhar Z Khan

Compared with standard lead positioning without echo guidance, the echo-guided procedure more often positioned the lead at a location within the ventricle considered most effective for pacing and most likely to correct dyssynchrony, Dr Fakhar Z Khan (Cambridge University, UK) explained to heartwire . The optimal lead location, he said, appears to be away from scar tissue and, more important, at the site of myocardial contraction occurring latest in the cardiac cycle.

In the Targeted Left Ventricular Lead Placement to Guide Cardiac Resynchronisation Therapy (TARGET) trial, LV lead placement guided by speckle-tracking echo was associated with smaller LV end-systolic volumes (LVESV) after six months, a decline in rate of death or HF hospitalization, and improved NYHA functional class and six-minute-walk distances.

Khan, who presented the findings here at the American College of Cardiology 2011 Scientific Sessions/i2 Summit, observed that frequently suboptimal lead placement has long been thought partly responsible for the wide observed variation in patient responses to CRT. Up to 40% of patients who get CRT devices don't improve clinically, he said.

It's typical for CRT pacing leads to be placed in standard configurations, but "one of the biggest messages [of TARGET] is that there should be individualization," according to Khan. "I think it does support the idea that one site doesn't fit all, that there is great heterogeneity in what constitutes an optimal site for pacing."

Speckle tracking, a relatively new echo modality, can help differentiate scarred from functional myocardium by delineating patterns of myocardial strain, showing which muscle segments are contracting and when within the cardiac cycle. The segment that contracts the latest is considered the best pacing site for addressing dyssynchrony, according to Khan.

Dr Michael Crawford

During the panel discussion that followed Khan's presentation of TARGET, Dr Michael Crawford (University of California, San Francisco), who wasn't connected to the trial, observed that the magnitude of LVESV and LVEF improvement with guided lead placement "was rather modest, yet the clinical effect was dramatic."

On that point, Dr Prakash C Deedwania (Veterans Affairs Central California Healthcare System and University of California San Francisco, Fresno) noted that there was no suggestion in the study that the structural/functional improvements were actually related to the clinical gains. "I think that's the gap that's remaining. I think clearly we can show echocardiographic improvement, and we can separately show clinical outcomes improvement, but are they the same patients?"

Hemodynamic data from the first 24 hours of CRT treatment, not included in his presentation, Khan replied, "show that as a result of the targeting process, you do get acute improvements in cardiac output."

The trial randomized 220 patients at two centers in the UK who were slated for CRT and had the standard indications, including heart failure, LVEF <35%, and QRS prolongation, to lead placement guided by speckle-tracking echo or conventional placement. The analysis, however, included only 207 patients due to some deaths, lead-placement failures, and losses to follow-up.

The randomization groups were similar at baseline with respect to chamber dimensions, medical therapy, ECG features, prevalence of diabetes, and NYHA functional class.

Dr Prakash C Deedwania

All patients underwent the echo procedure to identify an optimal pacing site, but in the control group the leads were positioned with blinding to the echo data.

They were all categorized according to whether their operator's effort ultimately positioned the LV lead at the echo-defined optimal site ("concordant" positioning) or whether the lead ended up one segment away from the optimal site ("adjacent" positioning) or at least two segments away ("remote" positioning).

Operators who used echo guidance were able to place the LV lead in an optimal position (as indicated by the echo data) significantly more often than those who didn't have echo guidance and significantly less often placed it in a remote location (p=0.011).

Positioning of LV Leads Relative to Echo-Defined Optimal Pacing Site, Echo-Guided vs Control Groups*

Positioning relative to optimal site Echo-guided, n=103 (%) Control, n=104 (%)
Concordant 63 47
Adjacent 26 29
Remote 10 15

*p=0.011 for the difference between the two trends

Operators in the two groups were similarly likely to place the LV lead in a region of scar, which occurred in 8% of guided cases and 15% of control cases (p=0.133).

The primary end point, defined as at least a 15% reduction in LVESV at six months, was reached by 70% of echo-guided patients and 55% of controls (p=0.031). The former ended up with significantly higher LVEF (p=0.03).

The main secondary end point, at least a one-step improvement in NYHA class, was reached by 83% of patients in the guided-placement group and 65% of controls (p=0.003). The former also showed significantly improved six-minute-walk distance (p=0.01) and quality-of-life scores (p=0.02).

There was no significant difference in all-cause mortality, but patients in the guided group showed a significant advantage in rate of death or HF hospitalization (p=0.03).

Compared with remote lead positioning, concordant positioning was associated with significantly lower all-cause mortality (p=0.02) and death or HF hospitalization (p<0.001).

Similarly, positioning of the lead in scar was associated with increased all-cause mortality (p=0.02) and death or HF hospitalization (p=0.009) compared with positioning away from scar.

When interviewed, Khan said the speckle-tracking echo technique is practical for wide use. His center has several years of experience with it and has gotten good at it, but it can be easily adopted elsewhere "with the appropriate training," he said.

"What we haven't done is preselect patients using criteria other than the standard criteria based on QRS duration. If we started using dyssynchrony parameters to select out those patients in whom we know the response isn't going to be good--because their myocardium is too extensively damaged, for example, or there's no dyssynchrony to start with--we may see even better results. That's certainly a study that should be done."

Kahn, Deedwania, and Crawford had no disclosures.