New England Journal of Medicine September 14, 2023

Ferric Carboxymaltose in Heart Failure with Iron Deficiency

Mentz RJ, Garg J, Rockhold FW, et al.

Bottom Line

In ambulatory patients with heart failure with reduced ejection fraction and iron deficiency, treatment with intravenous ferric carboxymaltose did not reach the prespecified significance level for improving a hierarchical composite of death, heart failure hospitalizations, and 6-minute walk distance compared to placebo.

Key Findings

1. The primary hierarchical composite outcome of death, heart failure hospitalizations, and change in 6-minute walk distance did not meet the prespecified significance level of 0.01 (Wilcoxon-Mann-Whitney P = 0.02; unmatched win ratio, 1.10; 99% CI, 0.99 to 1.23).
2. Death by month 12 occurred in 131 patients (8.6%) in the ferric carboxymaltose group and 158 patients (10.3%) in the placebo group.
3. A total of 297 hospitalizations for heart failure occurred by month 12 in the ferric carboxymaltose group compared to 332 in the placebo group.
4. The mean change from baseline to 6 months in the 6-minute walk distance was 8±60 m in the ferric carboxymaltose group and 4±59 m in the placebo group.
5. Serious adverse events during the treatment period were similar between the two groups, occurring in 413 patients (27.0%) in the ferric carboxymaltose group and 401 patients (26.2%) in the placebo group.

Study Design

Design
RCT
Double-Blind
Sample
3,065
Patients
Duration
12 mo
Median
Setting
Multicenter, Multinational
Population Ambulatory patients with heart failure, a left ventricular ejection fraction of 40% or less, and iron deficiency.
Intervention Intravenous ferric carboxymaltose given every 6 months as needed on the basis of iron indexes and hemoglobin levels, in addition to standard therapy for heart failure.
Comparator Placebo given every 6 months as needed, in addition to standard therapy for heart failure.
Outcome A hierarchical composite of death within 12 months after randomization, hospitalizations for heart failure within 12 months, or change from baseline to 6 months in the 6-minute walk distance.

Study Limitations

The trial employed a highly stringent prespecified significance level (alpha = 0.01), meaning the observed trend toward benefit (P = 0.02) was formally considered non-significant.
The trial was conducted during the COVID-19 pandemic, which may have reduced the expected rates of heart failure hospitalizations and impacted the completion of 6-minute walk distance tests.
Enrolling a generally stable, ambulatory outpatient population likely resulted in lower baseline cardiovascular risk and lower event rates, potentially diluting the absolute benefit of the intervention compared to higher-risk, post-acute heart failure cohorts.

Clinical Significance

The HEART-FID trial provides critical, large-scale safety and efficacy data regarding intravenous iron repletion in HFrEF. Although it failed to show a definitive statistical difference for its primary hierarchical composite endpoint due to a strict significance threshold of 0.01, the numerical trends favored ferric carboxymaltose for mortality and hospitalizations. The trial robustly confirmed the safety of repeated intravenous ferric carboxymaltose dosing in a large ambulatory population, reassuring clinicians who prescribe it for symptomatic improvement, though it tempers expectations regarding robust reductions in hard clinical events.

Historical Context

Iron deficiency is highly prevalent in patients with heart failure and is independently associated with diminished functional capacity and increased mortality, irrespective of anemia. Previous trials such as FAIR-HF, CONFIRM-HF, and AFFIRM-AHF established that intravenous ferric carboxymaltose improves symptoms, quality of life, and reduces heart failure hospitalizations, particularly after an acute decompensation. HEART-FID was designed as the largest definitive trial to date to assess whether these functional benefits translate into robust improvements in hard clinical endpoints (death and hospitalizations) in a broader, stable ambulatory HFrEF population.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Iron deficiency in heart failure often occurs in the absence of clinical anemia. What is the pathophysiological mechanism linking isolated iron deficiency to impaired myocardial and skeletal muscle function in HFrEF?

Key Response

This tests the foundational concept that iron is essential for cellular energy metabolism. Beyond erythropoiesis, iron is a critical cofactor for mitochondrial oxidative phosphorylation, cytochrome enzymes, and myoglobin. Its depletion directly impairs ATP production in both the myocardium and skeletal muscle, explaining why replacing it could theoretically improve exercise capacity and symptoms independently of hemoglobin levels.

Resident
Resident

Given the 'negative' primary outcome of the HEART-FID trial, how should clinicians approach the diagnosis and management of iron deficiency in ambulatory patients with HFrEF? Should intravenous iron be abandoned in this population?

Key Response

Residents need to integrate new trial data with established practices. While HEART-FID missed its prespecified significance level for the composite endpoint, the diagnostic criteria for iron deficiency (ferritin < 100 ng/mL, or 100-299 ng/mL with TSAT < 20%) remain critical. Prior trials (like AFFIRM-AHF) still support IV iron for symptom improvement and reducing hospitalizations. Therefore, IV iron should not be abandoned, but the decision should involve shared decision-making, emphasizing modest benefits in stable ambulatory patients.

Fellow
Fellow

The HEART-FID trial failed to meet its primary hierarchical endpoint, which appears to contrast with the results of AFFIRM-AHF and CONFIRM-HF. How do differences in the enrolled patient populations, particularly regarding the acuity of heart failure, explain these disparate findings?

Key Response

Fellows must critically compare trial populations. AFFIRM-AHF enrolled patients shortly after an acute heart failure hospitalization, a period of extremely high risk for readmission. In contrast, HEART-FID enrolled a stable, ambulatory chronic HF cohort. The lower baseline event rate in the ambulatory population makes it mathematically harder to detect a statistically significant reduction in hard outcomes like mortality and hospitalizations within the trial's timeframe.

Attending
Attending

HEART-FID utilized a hierarchical composite endpoint analyzed via the Finkelstein-Schoenfeld (win ratio) method. How does relying on this specific statistical approach impact clinical decision-making when the overall result is 'negative' but individual components, such as heart failure hospitalizations, trend in a positive direction?

Key Response

Attendings must interpret complex trial designs to guide practice. The hierarchical method prioritizes death, then hospitalization, then 6-minute walk distance. A non-significant overall win ratio means the drug did not definitively 'win' across this rigid hierarchy. Clinicians must weigh whether non-statistically significant but positive trends in morbidity (hospitalizations) justify the logistical burden, cost, and infusion center resources required for IV iron administration.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The primary analysis of HEART-FID relied on an unmatched win ratio with a stringent prespecified significance level (alpha=0.01) for the pivotal outcome. What are the methodological limitations of using hierarchical composites in cohorts with lower-than-expected mortality event rates, and how does this affect statistical power and Type II error?

Key Response

In a hierarchical composite, if the top tiers (death, hospitalization) have low event rates, the statistical test pushes most of the 'ties' to the continuous variable at the bottom (6MWD). If the trial was powered expecting higher upper-tier events, the shift compromises statistical power, making the overall win ratio highly sensitive to baseline risk and increasing the likelihood of a Type II error in lower-risk ambulatory cohorts.

Journal Editor
Journal Editor

Because intravenous iron has a distinct dark color and can transiently alter systemic iron biomarkers, maintaining strict blinding is notoriously difficult. As a peer reviewer, how would you evaluate the risk of performance and detection bias in the HEART-FID trial, particularly regarding the effort-dependent 6-minute walk distance component of the primary endpoint?

Key Response

A critical appraisal must scrutinize blinding integrity. If patients or assessors become aware of the treatment assignment due to visual cues of the infusion, skin staining, or unblinded labs, it could artificially inflate the effort-dependent 6-minute walk distance. A reviewer should look closely at the blinding protocols, who conducted the walking tests, and how missing 6MWD data (especially during the COVID-19 pandemic) was imputed.

Guideline Committee
Guideline Committee

Current ACC/AHA and ESC guidelines provide a Class IIa recommendation for intravenous iron in HFrEF with iron deficiency to improve symptoms and reduce hospitalizations. Does the failure of HEART-FID to meet its primary composite endpoint warrant downgrading this recommendation to Class IIb, or does the totality of evidence, including meta-analyses, sustain the current guidance?

Key Response

Guideline committees must synthesize all available data. While HEART-FID was technically a neutral trial, meta-analyses combining HEART-FID with FAIR-HF, CONFIRM-HF, and AFFIRM-AHF continue to demonstrate a significant reduction in recurrent heart failure hospitalizations and improvements in quality of life. The committee would likely maintain the Class IIa recommendation, concluding that the totality of evidence supports efficacy, albeit with perhaps a more modest absolute risk reduction in strictly ambulatory populations.

Clinical Landscape

Noteworthy Related Trials

2009

FAIR-HF Trial

n = 459 · NEJM

Tested

Intravenous ferric carboxymaltose

Population

Heart failure patients (NYHA class II or III) with iron deficiency

Comparator

Placebo

Endpoint

Self-reported Patient Global Assessment and NYHA functional class at week 24

Key result: Treatment with ferric carboxymaltose significantly improved self-reported symptoms, NYHA class, and exercise capacity compared to placebo.
2014

CONFIRM-HF Trial

n = 304 · Eur Heart J

Tested

Intravenous ferric carboxymaltose

Population

Symptomatic heart failure patients (LVEF <= 45%) with iron deficiency

Comparator

Placebo

Endpoint

Change in 6-minute walk test distance at week 24

Key result: Ferric carboxymaltose significantly improved functional capacity, symptoms, and quality of life over 1 year, with a signal indicating reduced hospitalizations for worsening heart failure.
2020

AFFIRM-AHF Trial

n = 1,132 · Lancet

Tested

Intravenous ferric carboxymaltose

Population

Patients stabilized after an acute heart failure episode with iron deficiency and LVEF < 50%

Comparator

Placebo

Endpoint

Composite of total heart failure hospitalizations and cardiovascular death up to 52 weeks

Key result: Ferric carboxymaltose reduced the risk of heart failure hospitalizations but did not have a statistically significant effect on cardiovascular death.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis