Who Should Get CRT? Resynchronization Benefits Seen in Heart Failure With LVEF >35%

September 14, 2009

September 14, 2009 (Boston, Massachusetts)Cardiac resynchronization therapy (CRT) can improve symptom status and shrink ventricular volumes in patients with LV ejection fractions higher than the 35% cut-off that, along with a wide QRS interval and advanced symptoms, now defines eligibility for CRT in the guidelines, suggests a retrospective analysis of data from a prospective international trial [1].

Along with the recent large prospective trials MADIT-CRT and REVERSE, and some smaller studies, the analysis suggests that "the current indications for CRT may not cover the whole gamut of patients who might benefit, and that we should explore other populations, for example, those with higher ejection fractions or those with narrow QRS," Dr Eugene S Chung (The Christ Hospital and The Ohio Heart and Vascular Center, Cincinnati) told heartwire .

"If we could more accurately identify who might benefit, how great would that be for them--to have a therapy that wasn't otherwise available to them?" The analysis suggests CRT therapy might be better guided by metrics other than LVEF, ones "that might predict whether someone is going to respond favorably. Whether that's other echo characteristics or the QRS interval, that remains to be seen. But it may be that ejection fraction should not be among the main entry criteria." Chung presented the study here today at the Heart Failure Society of America 2009 Scientific Meeting.

He and his colleagues retrospectively analyzed the relation between baseline LVEF and the prospectively defined end points from the randomized Predictors of Response to CRT (PROSPECT) study, which entered 426 patients in NYHA functional class 3-4 on optimal medical therapy, a QRS duration of >130 msec, and an LVEF <35% as assessed by echocardiography at each enrolling center.

The nonrandomized study's primary purpose was to evaluate a series of potential echocardiographic measures of ventricular dysynchrony as predictors of symptomatic and ventricular structural response to CRT at centers around the world, in the hopes of using some of them to sharpen the selection of CRT candidates likely to respond to the therapy. As previously reported by heartwire , none of the evaluated echo criteria--which include pulsed-Doppler and tissue-Doppler indices--added greatly to conventional echo criteria in predicting responses to CRT, at least as those proposed metrics were assessed under real clinical-practice conditions around the world.

The echo studies of LV function that determined enrollment eligibility were conducted at each participating center, and separate prospectively mandated quantitative evaluations of LVEF at the study's three echo core laboratories were available for 361 patients.

As it happened, LVEF was frequently underestimated at the enrolling centers, relative to core-lab evaluations, across the range of ejection fractions (not just for those near the 35% cut-off in the enrollment criteria).

Chung et al looked at the 86 of those patients, a substantial 24%, whose baseline core-lab LVEF exceeded 35% (mean, 43%), and compared them with the remaining 275 patients who had lower core-lab LVEFs (mean, 25%).

The mean QRS durations in the >35% and <35% LVEF groups were 157 msec and 165 msec, respectively.

Over six months, the two groups responded similarly to CRT with respect to the primary end points of LV end-systolic volume and a clinical composite score that accounted for independently adjudicated survival, hospitalization, heart failure symptom status, and NYHA functional class.

Clinical Composite Score and LVESV, Rates of Response to CRT (% of Patients) at Six Months by LVEF Group

End points LVEF<35 (%) LVEF>35 (%)
Clinical composite score response    
Improved 70.2 62.8
Unchanged 12.7 18.6
Worsened 17.1 18.6
LVESV response    
>15% decrease 57.8 50.8
Unchanged 34.5 34.9
>15% increase 7.6 14.3
All differences between LVEF groups NS.

LVESV=left ventricular end-systolic volume

Nor were significant differences between the LVEF groups seen in a range of secondary end points, including six-minute walk distance improvement of at least 10%, at least a nine-point improvement in Minnesota Living with Heart Failure quality-of-life score, >15% improvement in LV end-diastolic volume, any reduction in LV end-systolic or end-diastolic diameters or in LV mass, and any improvement in mitral regurgitation.

The mean LVEF improvement was more robust in the <35% group--a 33.6% relative improvement, compared with only a 3.2% relative improvement in the >35% group (p<0.0001), according to the researchers. Still, only 11.6% of the >35% group "failed to show any improvement in a primary or secondary outcome measure."

The analysis, Chung observed, raises the question of whether ECG or echo measures of dysynchrony might be worth targeting with CRT in patients with LVEF >35%. That should be explored in prospective trials, he said. The analysis, along with other completed and ongoing studies, questions all three of the main CRT eligibility criteria in the guidelines: an ejection fraction <35%, NYHA class 3-4, and a QRS duration >120 msec.

"The MADIT-CRT and REVERSE trials looked at patients with NYHA class 1-2 heart failure, and so they addressed one of the criteria," Chung noted. Patients with mild disease in those two trials showed significant symptomatic and reverse-remodeling gains with CRT, as previously reported by heartwire . "RETHINQ addressed and the currently enrolling ECHO-CRT trial is addressing CRT in patients with narrow QRS intervals. So the only one of the three major criteria that hasn't been prospectively addressed in multicenter trials is ejection fraction," Chung said. "Our small pilot analysis suggests that such a trial may be warranted."

PROSPECT was funded by Medtronic, from which Chung discloses receiving research grants and honoraria for speaking.