Gains in HF Drug Therapy Raise Questions About ICD Benefit in Modern Era

SAN FRANCISCO, CA—A primary prevention implantable cardioverter-defibrillator (ICD) seems to lessen the risk of dying among patients with heart failure and reduced ejection fraction (HFrEF) even on the background of contemporary therapy that includes a sodium-glucose cotransporter 2 (SGLT2) inhibitor, an analysis of the EMPEROR-Reduced trial suggests.

Among patients treated with empagliflozin (Jardiance; Boehringer Ingelheim/Eli Lilly), those with versus without an ICD at baseline had lower risks of all-cause death (8.5 vs 12.7 per 100 patient-years) and sudden cardiac death (1.6 vs 3.3 per 100 patient-years) over 2 years of follow-up.

Both differences, however, fell shy of statistical significance, Mehmet Aktas, MD (University of Rochester, NY), reported here at Heart Rhythm 2022.

“The addition of SGLT2 inhibitors to the cavalry of pharmacological therapy available for heart failure patients has attenuated mortality outcomes, but may not have obviated the need for a primary prevention ICD,” he concluded, calling for further research.

“In order to better answer this question,” Aktas said, “a large-scale randomized controlled trial with randomization of patients to ICD and no-ICD groups, with all participants on optimal guideline-directed medical therapy, is necessary to evaluate the benefit of an ICD in conjunction with contemporary guideline-directed medical therapy.”

He noted that the trials demonstrating a reduction in mortality with primary prevention ICDs were performed more than two decades ago at a time when guideline-directed medical therapy included ACE inhibitors/ARBs, beta-blockers, and—in a small proportion of patients—mineralocorticoid receptor antagonists (MRAs).

But that has changed, with both sacubitril/valsartan (Entresto)—an angiotensin receptor-neprilysin inhibitor (ARNI)—and the SGLT2 inhibitors now recommended as key parts of HF therapy in the most recent US and European guidelines. These agents have further improved outcomes among HFrEF patients. In EMPEROR-Reduced, for example, empagliflozin cut the risks of CV death or hospitalization for worsening HF (19.4% vs 24.7%; HR 0.75; 95% CI 0.65-0.86) relative to placebo.

With better medical therapy, it’s unclear whether primary prevention ICDs will have the same benefits as seen in the older trials.

The addition of SGLT2 inhibitors to the cavalry of pharmacological therapy available for heart failure patients has attenuated mortality outcomes, but may not have obviated the need for a primary prevention ICD. Mehmet Aktas

To explore that question, Aktas and his colleagues turned to the empagliflozin-treated patients from EMPEROR-Reduced, using propensity-score matching to compare 535 patients with an ICD at enrollment with an equal number of patients without an ICD. Matching was based on age, sex, LVEF, body mass index, NYHA class, and etiology of heart failure (ischemic or nonischemic).

Mean patient age was 68, and 81% were men. More patients (73%) had NYHA class II disease. Those with an ICD were less likely to have diabetes (44% vs 56%) and hypertension (70% vs 75%), and more likely to have hypercholesterolemia (68% vs 62%) and atrial fibrillation (44% vs 30%).

Differences were seen in baseline medical therapy. The ICD group had greater use of beta-blockers (97% vs 93%) and ARNIs (26% vs 14%) but less use of ACE inhibitors/ARBs (64% vs 75%) and MRAs (66% vs 68%).

Comparing the ICD and no-ICD groups, there is a separation in the curves for both all-cause mortality (HR 0.74; 95% CI 0.51-1.07) and sudden cardiac death (HR 0.59; 95% CI 0.31-1.15) in favor of the patients with ICDs, although neither relationship reached statistical significance. For a composite of HF hospitalization or CV death, there was no hint of a difference between the two groups (HR 1.02; 95% CI 0.76-1.38).

We’ve come a long way in 20, 30 years . . . in helping people survive longer even without an ICD. Julia Indik

Discussing the results after Aktas’ presentation, Julia Indik, MD, PhD (University of Arizona College of Medicine – Tucson), one of the session’s moderators, pointed out that it’s unknown how many of the patients in the analysis were being treated with cardiac resynchronization therapy (CRT). “That could confuse the analysis, analogous to what’s been said about the DANISH trial [regarding] the influence of CRT therapy and its beneficial effect in the control group.”

Heather Bloom, MD (Emory University School of Medicine and Atlanta VA Medical Center, GA), the other moderator, said what stood out for her from this study and others like it was that the results are driven by patients with NYHA class II heart failure. “I always wonder how that affects the class IIIs and IVs that we really don’t have much data [on],” she said. She noted, too, that “even in this very well-controlled trial, there were still a lot of people that weren’t on goal-directed medical therapy.”

Still, Bloom said, the analysis generates a good hypothesis for future research.

Taking a broader view of the study, Indik added that “it also tells us, at least, we’ve come a long way in 20, 30 years . . . in helping people survive longer even without an ICD.”