A study subgroup with extended follow-up
The new, extended follow-up analysis presented by Dr. Sapp included 1,050 of the original 1,798 patients (58%) enrolled at any of eight participating Canadian centers that each enrolled at least 100 patients and followed them through the end of 2021 (the full study cohort came from 34 centers, including 10 centers outside Canada). This subgroup included 520 patients randomized to receive CRT-D and 530 who received an ICD. Although this was a post hoc subgroup analysis, the CRT-D and ICD arms matched closely in all measured baseline characteristics.
The prespecified primary outcome of this follow-up analysis was the rate of all-cause mortality. Because of their longer disease trajectory, this pared-down study cohort included many more patients with NYHA class II function, 803, and in this subgroup CRT-D exerted a significant 23% incremental reduction in mortality compared with ICD treatment. CRT-D also produced a 17% relative reduction in long-term mortality among patients with NYHA class III function at baseline, but this point estimate of relative benefit was not significant in this subgroup of just 247 patients, said Dr. Sapp, a cardiologist and professor at Dalhousie University & Nova Scotia Health in Halifax.
Based on the original RAFT results from 2010, as well as on evidence from several other trials, the current heart failure management guideline from the AHA, the American College of Cardiology, and the Heart Failure Society of America give the highest level of recommendation, level 1, for CRT in patients with a left ventricular ejection fraction of 35% or less, sinus rhythm with left bundle branch block, a QRS duration of at least 150 msec, and NYHA class II, III, or ambulatory IV symptoms while on guideline-directed medical therapy.
The guideline also gives class 2a (“can be useful”) or 2b (“may be considered”) recommendation for certain other heart failure patients, including those with a QRS duration of 120-149 msec, a left ventricular ejection fraction as high as 50%, no left bundle branch block, or NYHA class I symptoms.
Don’t wait to start CRT
Although this 2022 guideline, as well as earlier versions that had roughly similar recommendations for CRT for about a decade, have led to “common” use of CRT in appropriate patients in U.S. practice, “it has not been used as much as it should be, in part because there’s been a feeling that CRT mostly treats symptoms and so perhaps you can wait” to start it, said Dr. Stevenson.
The findings from the new, extended follow-up RAFT analysis give increased urgency to starting CRT “as soon as possible” in appropriate patients with heart failure, even before they stabilize on guideline-directed medical therapy, said Dr. Stevenson. She also downplayed any ambiguity in the RAFT findings about optimal medical therapy, which during the RAFT study included traditional triple therapy at a time before treatment with sacubitril/valsartan (Entresto) and sodium-glucose cotransporter 2 (SGLT2) inhibitors became recommended.
“There is no reason to think that these treatments will negate the benefit of CRT for patients with heart failure with reduced ejection fraction and a wide left bundle branch block,” Dr. Stevenson said.
She also believes that the extended follow-up results, which showed clear efficacy for CRT-D in patients with NYHA class II function, support the case for upgrading the current 2b recommendation for using CRT treatment in patients with NYHA class I function and ischemic heart failure to a 2a recommendation regardless of whether or not patients have coronary artery disease. “The difference between class I and class II depends more on a patient’s lifestyle rather than on the severity of their heart failure,” Dr. Stevenson noted. “The RAFT study results encourage us to reexamine the clinical class and timing for CRT” in the current heart failure guideline.
RAFT received partial sponsorship from Medtronic. Dr. Sapp has been a consultant to Abbott, Biosense Webster, Medtronic, and Varian and has received research funding from Abbott and Biosense Webster. Dr. Stevenson had no disclosures.