A new article offers insight on the challenges of reducing mortality rates in patients with heart failure with mildly reduced ejection fraction (HFmrEF) and preserved ejection fraction (HFpEF). The paper also examined why randomized controlled trials (RCTs) have failed to demonstrate statistically significant mortality reductions for these patients, compared to notable successes in heart failure with reduced ejection fraction (HFrEF). Commenting on this observation, led by author Toru Kondo, MD, PhD, and colleagues, said, “this probably reflects the statistical power of trials to date to show an effect on mortality rather than mechanistic differences between HFmEF/HFpEF and HFrEF or differences in treatment efficacy.” “It is probably not practical or affordable to design trials in HFmrEF/HFpEF to show a reduction in all-cause mortality unless the treatment tested also reduces noncardiovascular causes of death,” the paper’s authors concluded.” Statistical significance Dr. Kondo, who is from the British Heart Foundation Cardiovascular Research Centre at the University of Glasgow, Scotland, led the discussion by acknowledging recent treatment breakthroughs that have reduced mortality in HFrEF, but reserved judgement for HFmrEF and HFpEF outcomes. "To date, no randomized controlled trial has demonstrated a statistically significant reduction in mortality, or cardiovascular mortality, in patients with heart failure and HFmrEF or heart failure and HFpEF. “[This is] in notable contrast to the beneficial effect of many treatments on these outcomes in patients with HFrEF." To further demonstrate the impact of mechanistic obstacles when compared to statistical barriers the authors state that the failure to reduce mortality in HFmrEF/HFpEF does not originate from fundamental biological differences in disease mechanisms compared to HFrEF. On the contrary, they add that this gap forms as a result of insufficient statistical power in existing trials citing the example of treatments such as SGLT2 inhibitors, which have shown consistent benefits across heart failure types. In addition, the authors also point out that the trials in HFmrEF/HFpEF often lack the large sample sizes needed to detect modest reductions in cardiovascular mortality. Meanwhile, in trials looking at HFrEF, the reduction in mortality is simpler to monitor as cardiovascular deaths make up to 85% of all fatalities. Conversely, around 40-60% of deaths in patients with HFmrEF/HFpEF are cardiovascular, accelerating the need for larger trials to spot smaller treatment effects. Categorizing deaths Published in the November 26 issue of the Journal of the American College of Cardiology, the paper goes on to describe the importance of categorizing deaths from unknown or uncertain causes. The team point out that in the past, RCTs have classified these deaths as either cardiovascular, non-cardiovascular, or excluded them altogether. However, more contemporary HFmrEF/HFpEF trials, have followed a more conservative approach, which has led to a smaller number of deaths classified as cardiovascular. While this method improves accuracy, it lowers the number of modifiable events available for analysis, further reducing statistical power. Linked to this issue, the paper also highlights the substantial resources required to conduct sufficiently powered trials for HFmrEF/HFpEF. “…a sample size of approximately 15,000 patients with a median follow-up of 2 years would be required to demonstrate a 15% relative risk reduction in cardiovascular mortality with 80% power, assuming a 4% annual cardiovascular mortality rate (and overall 2-sided alpha 5%),” said the paper, which is also published Monday online. The authors added that this is especially because the median follow-up would need to be twice as long as in typical heart failure trials, “(i.e., more like 5-6 years than 2-3 years). Mutually beneficial Despite the paper’s conclusions, the authors highlight that effective treatments for HFrEF should also benefit HFmrEF/HFpEF patients in theory. This is especially relevant for modifiable cardiovascular deaths caused by worsening heart failure or sudden cardiac events. However, as the authors conclude, proving this requires a significant change in trial design, with an emphasis on larger-scale investigations or cutting-edge approaches with the ability to circumvent statistical limitations. Additionally, the authors also urge further work in re-evaluating endpoints in heart failure trials, commenting that composite outcomes that include hospitalizations, quality of life, and functional measures alongside mortality could better capture treatment benefits and guide clinical practice. Sources: Kondo T, Henderson AD, Docherty KF, et al. Why Have We Not Been Able to Demonstrate Reduced Mortality in Patients With HFmrEF/HFpEF? J Am Coll Cardiol. 2024;84: :2233–2240 Image Credit: REDPIXEL – stock.adobe.com